What Are Optimized Analytics?

What Are Optimized Analytics?

There is a lot of buzz around big data and data analytics. Typically these terms imply that someone, or some computer, is trying to detect patterns in, or creating information from, a company’s data. There are a lot of definitions floating around, but none that truly capture the integrated approach of Optimized Analytics. This blog post provides further definition and clarity around Optimized Analytics. 

Optimized Analytics – Approachable analytics that provide an actionable plan to achieve a desired business outcome in the most effective way.

In the world of data science there are descriptive, predictive, and prescriptive analytic tools and techniques. Descriptive analytics techniques are used to analyze what happened to your business in the past. They provide hindsight. Predictive analytics tools are good for identifying what is going on in your business today, and for making forecasts about tomorrow. These tools provide insight. Prescriptive analytics techniques, like optimization, offer a path forward. These techniques tell you what actions to take to achieve a desired outcome. This is often thought of as business foresight.

The problem with these definitions is that they imply that each technique can be done independent of the others. This is a bad assumption. In truth, each of these techniques build on one another. 

Your data is the raw material – like stones. When you organize your data with descriptive analytics it builds a foundation upon which to build your decisions. The foundation represents a baseline for your business. It describes what you have done to date and determines any trends you can glean from your data. 

Once you have a solid foundation in place you can build another layer with predictive analytics. In this layer, you use the findings from your descriptive techniques to help make predictions about the future of your business. Typically, this means using forecasting, simulation, or machine learning techniques. 

When you have a prediction about the future state of your business it is great for planning. The problem is, these techniques do not give you an actionable plan to achieve a desired outcome. Prescriptive analytics are the capstone of the pyramid. By using optimization techniques, a business can build the best plan to achieve their goals. Like any structure, however, you cannot add the capstone without first building the foundation. Optimization is no different. You cannot build the best plan for the future without using the descriptive and predictive analytic outcomes as inputs to the optimization.

 A recent Gartner Report echoes these thoughts by highlighting the increasing complexity of business decision and suggesting we need a combination of both predictive and prescriptive analytics to solve them. They suggest that predictive outcomes must flow into prescriptive models. To extend it further, we contend it is the harmonization of descriptive, predictive, and prescriptive analytics to produce a desired business outcome that create Optimized Analytics.

In addition, to achieve a desired business outcome, Optimized Analytics must be both actionable and approachable. Actionable means that the solution provides specific actions for the organization to take in order to best achieve their desired outcome. Approachable implies that the average business user can arrive at the best solution with little help from a data scientist. Optimized analytics should be delivered to the average business user within an easy-to-use framework. 

Lastly, Optimized Analytics are not a one-time event. They provide a process for continued improvement. Once you determine an optimal plan, based on your desired business outcome, you must evaluate how you are doing against that plan. This is done by continuing to analyze new data as it becomes available. If it indicates a shift in your business, you need to create new predictions and use them to generate a new optimized plan. By adopting a continuous process, Optimized Analytics provide a path forward that will remain relevant to your business.

 Now that you understand Optimized Analytics, it is time to put them to work for your business. At ORM Technologies we specialize in applying Optimized Analytics to the sales and marketing domain. If you would like to know more about what we do, or have questions, please let us know at [email protected].

**This article was originally posted on ORM Technologies' Blog**?

Scott Nestler

Decision Scientist

7 年

Thanks for sharing your thoughts with the group. I agree with much of what you say in this post. However, it isn't clear to me why you believe the terms descriptive, predictive, and prescriptive indicate that they are performed separately from one another. A real-world project often includes using a variety of techniques across the analytic continuum. I'm not sure that what you are calling optimized analytics is any different from what others would call prescriptive analytics that is informed by analysis of inputs, sensitivity of results to changes in inputs, forecasts, and assumptions, etc. Please let me know what I am missing.

要查看或添加评论,请登录

Pete Furseth的更多文章

社区洞察

其他会员也浏览了